Micronics SLS 3D Printer Guide

• 1
  How the powder bed actually builds a part
• 2
  Powder choice drives strength, heat resistance and cost
• 3
  What SLS handles well and what it does not
• 4
  From grey build cake to a usable part
• 5
  Where SLS stops being the economical answer
• 6
  SLS versus CNC machining: choosing by requirement

A Micronics SLS machine holds a powder bed at a temperature just below the melting point of the polymer. A roller or blade spreads a layer of powder, typically 0.08 to 0.12 mm thick, across the build area. A CO2 laser then scans the cross-section for that layer and raises the polymer to its melting range. The melted particles fuse, the piston drops by one layer, and the cycle repeats.

Because the surrounding powder is already hot, the fused layer cools slowly and evenly. That is the reason SLS parts are less warped than parts made on a machine that pulls hot plastic through a nozzle into room-temperature air. It is also why the build chamber is the real process parameter. Bed temperature, laser power, scan speed and layer thickness are set together, not one at a time.

The powder around the part is the support. No separate support structure is generated, so there is nothing to cut off a downward-facing surface and nothing to sand out of an internal channel. Undercuts, deep pockets and nested features are all buildable as long as loose powder can be removed after the build.

That last condition is the trap. A closed internal cavity with no exit will stay full of powder. So will a blind slot narrower than the powder can flow through. Design for powder removal first, and the rest of the process becomes predictable.

• 1
  Layer thickness0.08–0.12 mm is the common band; thinner layers cost build time
• 2
  Thermal fieldUniform bed temperature is what limits warp, not the laser alone
• 3
  SupportUnfused powder supports the part; no breakaway structures
• 4
  RemovalEvery internal void needs a powder escape path

Unfilled PA12 is the default. It gives a good balance of tensile strength, impact resistance and dimensional stability, and it machines and taps reasonably well. Parts come out with a slightly porous, matte surface. If you need a smooth or sealed surface, plan a finishing step from the start rather than hoping the printer delivers it.

Glass-filled PA12 (PA12-GF) raises stiffness and heat resistance. Filled grades can hold useful mechanical properties up to roughly 170 °C, which covers many under-hood brackets and pump housings. The trade is brittleness and a rougher surface, plus faster wear on the machine's recoating system. For higher temperatures, polymer powder is the wrong tool. Inconel and titanium parts belong in a metal process, not in an SLS powder bed.

Elastomeric powders such as TPU open another lane: gaskets, grips and vibration pads that would be difficult to mold in low volume. They stretch, they damp, and they do not hold tight tolerances. Do not specify a bearing bore in a flexible material and expect a press fit.

Every powder grade has a reuse policy. A fraction of the powder in the bed is freshly melted each cycle and a fraction is reclaimed and blended. Reuse ratio affects color and mechanical consistency from build to build. If your part is cosmetic or structural, ask what the reuse ratio is and keep it fixed across the program.

SLS is strong on organic shapes: ducting with sweeping bends, lattice-filled panels, brackets with material only where the load path runs, and housings that combine several machined parts into one. Because there is no tooling and no support removal, the cost of complexity is low. Adding ribs and bosses usually adds almost nothing to the price.

It is weak on anything that depends on a precise fit. Expect roughly ±0.3 mm on a well-controlled build, and worse on long thin features that cool unevenly. A Ø6 H7 bore is not an SLS feature. Neither is a flat sealing face or a thread that must take load without an insert. Design those features oversized and cut them afterward, or plan for a machined insert.

Wall thickness matters more than people expect. Below about 0.8 mm, thin walls can curl or build inconsistently. Above about 4 mm, thick sections trap heat and can show porosity or dimensional drift in the core. A 2 to 3 mm wall with ribs is a reliable starting point for most functional parts.

Small holes are another boundary. Holes under about 1 mm tend to close up or come out undersized, so drill them after the build. Vertical holes hold their size better than holes that sit flat on a layer, because the circular cross-section is defined by the laser path rather than by the layer stack.

• 1
  Good fitDucts, brackets, lattice panels, consolidated housings
• 2
  Poor fitPrecision bores, sealing faces, load-bearing threads
• 3
  Wall rangeAbout 0.8–4 mm; 2–3 mm with ribs is safe
• 4
  Tiny holesUnder 1 mm should be drilled after the build

A fresh build comes out as a cake of powder with parts embedded in it. The operator breaks the cake out, brushes off loose powder and, for most geometries, bead blasts the surface. Bead blasting does more than clean. It removes partially sintered powder that clings to the surface and would otherwise shed inside your assembly.

If the part needs a smooth finish, the options are tumbling, media polishing or a sealed coating. None of them turn SLS into an injection-molded surface. What they do is reduce the roughness enough that the part wipes clean and looks consistent from unit to unit. For fluid passages, a coating can also close the surface porosity so the part does not weep.

Dimensional features are usually finished by machining. SLS gives you the near-net shape, then a CNC operation cuts the bearing bores, sealing faces, threads and dowel holes. This hybrid route is common in low-volume production because it removes the risk of a sintered fit that is out of tolerance without paying for a full mold.

Tolerance on the machined features follows the CNC process, not the printer. GreatLight holds ±0.005 mm on machined features and Ra 0.8–1.6 μm on standard finishes, so a sintered blank with machined interfaces can meet a real drawing.

SLS wins on low volume and high complexity. One part or fifty parts cost roughly the same per unit, because there is no tooling. That is the whole argument for the process in prototyping and bridge production. It loses that advantage as volume climbs, because machine time and powder cost scale with part count.

The crossover depends on part size and geometry, but the pattern is consistent. Once a design is frozen and annual volume reaches the thousands, injection molding or die casting usually beats SLS on unit cost. Below that, SLS keeps the tooling money in your pocket.

There is a middle zone where CNC machining is the better call: simple prismatic parts, tight tolerances, metal, or a surface finish that has to be functional rather than cosmetic. Machining a bracket from 6061-T6 or 17-4PH gives you the strength and the tolerance in one step, with no post-machining operation. For a part with three holes and two flat faces, printing it first is wasted motion.

The practical rule: use SLS when geometry is the hard part. Use CNC when tolerance, material or surface is the hard part. Use both when you need a complex near-net shape with precise interfaces.